88 THE FLOWERING PROCESS 



commonly some plants in a given treatment will flower and others 

 will not, and results may be expressed as a percentage of flowering 

 plants compared to the total replications of a treatment. A refinement 

 of this approach is to record the percentage of plants which reach a 

 certain flowering stage within some arbitrary time limit. Or the time 

 required to reach a certain arbitrary stage may be recorded. It is 

 common, for example, to present data as the number of days to the 

 first visible bud. In such studies the rate of floral development is the 

 factor actually measured, but it is often assumed that this is a function 

 of the initial flowering stimulus produced by the act of induction. 

 While this may be true, environmental effects on development should 

 not be overlooked. 



Another procedure closely related to these is to examine plants at 

 some arbitrary time and assign arbitrary stages of floral development 

 to these plants. Various workers have used both macroscopic and 

 microscopic stages. A first problem arises in trying to assign the 

 stages; a second arises in knowing what relationship the numbers 

 assigned to the stages might bear to each other. In spite of these 

 problems we use this method, and our reasons for adopting it are 

 described in the next section. 



Some workers have measured the number of nodes to the first 

 flower, a convenient way of expressing data with certain species. The 

 height of the flowering stem may be measured in plants that send up 

 a flowering stalk from a rosette. The number of flowers or the 

 number of flowering nodes may be counted. Such data may be 

 compared to the number of leaves on the plant, allowing one to omit 

 the time factor, which is always present in the methods described in 

 the above paragraphs. WilHam Hillman in his book (3) on flowering 

 is quite concerned with this problem and suggests that we should 

 always base our results on some sort of leaf index. In certain studies 

 this can be quite important, and the expression of the data may have 

 considerable influence upon the interpretation of the results. In my 

 opinion, however, the time factor is seldom a complication when 

 control plants are used correctly, and leaf index methods often 

 involve waiting many weeks for the final data. If space is limited this 

 will reduce the number of experiments which can be performed, and 

 one must decide whether or not the advantages to be gained are 

 worth the cost in terms of time and ultimate eff'ort. 



